|Publication number||US7272454 B2|
|Application number||US 10/454,937|
|Publication date||Sep 18, 2007|
|Filing date||Jun 5, 2003|
|Priority date||Jun 5, 2003|
|Also published as||CN1598720A, CN1598720B, DE102004026979A1, US20040249483|
|Publication number||10454937, 454937, US 7272454 B2, US 7272454B2, US-B2-7272454, US7272454 B2, US7272454B2|
|Inventors||Wilhelm K. Wojsznis, Terrence L. Blevins, Ashish Mehta|
|Original Assignee||Fisher-Rosemount Systems, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (18), Non-Patent Citations (6), Referenced by (29), Classifications (13), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present invention relates generally to process control systems and, more particularly, to the use of advanced control blocks, such as model predictive control blocks or optimizers, in process control systems having non-linear response characteristics.
Process control systems, such as distributed or scalable process control systems like those used in chemical, petroleum or other processes, typically include one or more process controllers communicatively coupled to each other, to at least one host or operator workstation and to one or more field devices via analog, digital or combined analog/digital buses. The field devices, which may be, for example valves, valve positioners, switches and transmitters (e.g., temperature, pressure and flow rate sensors), perform functions within the process such as opening or closing valves and measuring process parameters. The process controller receives signals indicative of process measurements made by the field devices and/or other of information pertaining to the field devices, uses this information to implement a control routine and then generates control signals which are sent over the buses to the field devices to control the operation of the process. Information from the field devices and the controller is typically made available to one or more applications executed by the operator workstation to enable an operator to perform any desired function with respect to the process, such as viewing the current state of the process, modifying the operation of the process, etc.
In the past, conventional field devices were used to send and receive analog (e.g., 4 to 20 milliamp) signals to and from the process controller via an analog bus or analog lines. These 4 to 20 ma signals were limited in nature in that they were indicative of measurements made by the device or of control signals generated by the controller required to control the operation of the device. However, in the past decade or so, smart field devices including a microprocessor and a memory have become prevalent in the process control industry. In addition to performing a primary function within the process, smart field devices store data pertaining to the device, communicate with the controller and/or other devices in a digital or combined digital and analog format, and perform secondary tasks such as self-calibration, identification, diagnostics, etc. A number of standard and open smart device communication protocols such as the HART®, PROFIBUS®, WORLDFIP®, Device-Net®, and CAN protocols, have been developed to enable smart field devices made by different manufacturers to be used together within the same process control network. Moreover, the all-digital, two-wire bus protocol promulgated by the Fieldbus Foundation, known as the F
In any event, the process controllers (or field devices) are typically programmed to execute a different algorithm, sub-routine or control loop (which are all control routines) for each of a number of different loops defined for, or contained within a process, such as flow control loops, temperature control loops, pressure control loops, etc. Generally speaking, each such control loop includes one or more input blocks, such as an analog input (AI) function block, a single-output control block, such as a proportional-integral-derivative (PID) or a fuzzy logic control function block, and a single output block, such as an analog output (AO) function block. These control loops typically perform single-input/single-output control because the control block creates a single output used to control a single process input, such as a valve position, etc. However, in certain cases, the use of a number of independently operating, single-input/single-output control loops is not very effective because the process variables or process outputs being controlled are effected by more than a single process input and, in fact, each process input may effect the state of many process outputs. An example of this might occur in, for example, a process having a tank being filled by two input lines, and being emptied by a single output line, each line being controlled by a different valve, and in which the temperature, pressure and throughput of the tank are being controlled to be at or near desired values. As indicated above, the control of the throughput, the temperature and the pressure of the tank may be performed using a separate throughput control loop, a separate temperature control loop and a separate pressure control loop. However, in this situation, the operation of the temperature control loop in changing the setting of one of the input valves to control the temperature within the tank may cause the pressure within the tank to increase, which, for example, causes the pressure loop to open the outlet valve to decrease the pressure. This action may then cause the throughput control loop to close one of the input valves, thereby effecting the temperature and causing the temperature control loop to take some other action. As will be understood in this example, the single-input/single-output control loops may cause the process outputs (in this case, throughput, temperature and pressure) to oscillate without ever reaching a steady state condition, which is undesirable.
Model predictive control (MPC) or other types of advanced control have been used in the past to perform control in these types of situations. Generally speaking, model predictive control is a multiple-input/multiple output control strategy in which the effects of changing each of a number of process inputs on each of a number of process outputs is measured and these measured responses are then used to create a typically linear model of the process. The linear model of the process is inverted mathematically and is then used as a multiple-input/multiple-output controller to control the process outputs based on changes made to the process inputs. In some cases, the process model includes a process output response curve for each of the process inputs and these curves may be created based on a series of, for example, pseudo-random step changes delivered to each of the process inputs. These response curves can be used to model the process in known manners. Model predictive control is known in the art and, as a result, the specifics thereof will not be described herein. However, model predictive control is described generally in Qin, S. Joe and Thomas A. Badgwell, “An Overview of Industrial Model Predictive Control Technology,” AIChE Conference, 1996. Furthermore, U.S. Pat. No. 6,445,963, the disclosure of which is hereby expressly incorporated by reference herein, discloses a method of integrating a model predictive control block into a process control system for use in controlling a process.
While model predictive control blocks work well in standard processes with behaviors that can be generally modeled using linear techniques, it is difficult to use model predictive control blocks to control processes (or using process outputs) that exhibit non-linear behavior or that have a non-linear relationship with respect to the control inputs delivered to the process. In general, it is believed that MPC applications for non-linear processes would need to be adapted to provide non-linear optimization, to include non-linear controller generation and to use non-linear models instead of linear models within the MPC block. For example, it is known to provide non-linear modeling in series with a linear MPC technique to thereby adapt the inputs to the MPC block or the control outputs of the MPC block for non-linear processes. Such non-linear modeling may take the form of first principle models, high fidelity simulation, a mixture of non-linear gains and linear dynamics (Wiener or Hammerstein models) or neural network models. Additionally, it is known to adapt an MPC technique to include non-linear capability by using a neural network to recalculate the process model gains during each scan of the process control routine. However, gain recalculation is complex, involves calculation of derivatives from the process model resulting in additional modeling errors, can be computationally expensive and quite time consuming to implement. Thus, these methods of MPC adaptation are difficult and complex to implement within the MPC logic and may not be truly justified or practical in most process plant situations.
While these problems exist for model predictive controllers, the same or similar problems exist in the development and use of other advanced multiple-input/multiple output control blocks or systems, such as neural network controllers, multi-variable fuzzy logic controllers, real time optimizers, etc.
A multiple input/multiple-output control routine, such as a model predictive control routine, includes a generally linear process model that produces a prediction vector for one or more process outputs and a non-linear process model that produces a prediction signal, which may be a prediction vector or a steady state prediction value, to be used to correct the prediction vector for the one or more of the process outputs. The corrected prediction vector is then used in a normal manner to produce a set of control signals for controlling the process. The non-linear process model and the advanced control routine, such as the model predictive control routine, may be developed and operated in known and relatively simple manners which makes this system easy to implement and use when controlling a process having non-linear characteristics. Further, this system does not need to be modified to integrate a non-linear model into an multiple-input/multiple-output control block or to recalculate the controller gains used by the multiple-input/multiple-output control block. Furthermore, the non-linear process model is easy to develop (as it will generally reflect or model the non-linear nature of one output to one or more inputs) and may generally remain the same throughout operation of the process.
In one embodiment, a process controller adapted to be used to control a process having a set of process outputs effected by a set of process control input signals includes a multiple-input/multiple-output controller adapted to receive an indication of the process outputs to develop the set of process control input signals and a process model, which may be a non-linear process model, adapted to receive the set of process control input signals to produce a prediction signal for one of the process outputs. The multiple-input/multiple-output control element, which may be a controller, an optimizer, etc., includes another process model, which may be a standard linear process model, that develops a prediction vector for each of the process outputs and a correction unit adapted to modify the prediction vector for the one of the process outputs using the prediction signal for the one of the process outputs to thereby adjust for the non-linearities of the process.
Referring now to
The field devices 20-27 may be any types of devices, such as sensors, valves, transmitters, positioners, etc. while the I/O cards 28 and 29 may be any types of I/O devices conforming to any desired communication or controller protocol. In the embodiment illustrated in
The controller 12 includes a processor 12 a that implements or executes one or more process control routines, which may include control loops, stored in a computer readable memory 12 b therein or otherwise associated therewith and communicates with the devices 20-27, the host computers 16 and the data historian 14 to control a process in any desired manner. It should be noted that any control routines or elements described herein may have parts thereof implemented or executed by processors in different controllers or other devices, such as one or more of the field devices 24-27 if so desired. Likewise, the control routines or elements described herein to be implemented within the process control system 10 may take any form, including software, firmware, hardware, etc. For the purpose of this invention, a process control element can be any part or portion of a process control system including, for example, a routine, a block or a module stored on any computer readable medium. Control routines, which may be modules or any part of a control procedure such as a subroutine, parts of a subroutine (such as lines of code), etc. may be implemented in any desired software format, such as using ladder logic, sequential function charts, function block diagrams, or any other software programming language or design paradigm. Likewise, the control routines may be hard-coded into, for example, one or more EPROMs, EEPROMs, application specific integrated circuits (ASICs), or any other hardware or firmware elements. Still further, the control routines may be designed using any design tools, including graphical design tools or any other type of software/hardware/firmware programming or design tools. As a result, it will be understood that the controller 12 may be configured to implement a control strategy or control routine in any desired manner.
In one embodiment, the controller 12 implements a control strategy using what are commonly referred to as function blocks, wherein each function block is a part (e.g., a subroutine) of an overall control routine and operates in conjunction with other function blocks (via communications called links) to implement process control loops within the process control system 10. Function blocks typically perform one of an input function, such as that associated with a transmitter, a sensor or other process parameter measurement device, a control function, such as that associated with a control routine that performs PID, fuzzy logic, etc. control, or an output function which controls the operation of some device, such as a valve, to perform some physical function within the process control system 10. Of course, hybrid and other types of function blocks exist. Function blocks may be stored in and executed by the controller 12, which is typically the case when these function blocks are used for, or are associated with standard 4-20 ma devices and some types of smart field devices such as HART devices, or may be stored in and implemented by the field devices themselves, which can be the case with Fieldbus devices. While the description of the control system is provided herein using a function block control strategy, the control strategy or control loops or modules could also be implemented or designed using other conventions, such as ladder logic, sequential function charts, etc. or using any other desired programming language or paradigm.
As illustrated by the exploded block 30 of
It will be understood that the function blocks illustrated in
As illustrated in
While the non-linear process model 42 is illustrated in
As is known, the MPC block 44 includes a control prediction process model 70, which may be an N by M+D step response matrix (where N is the number of control variables CV (process outputs) plus the number of auxiliary variables AV, M is the number of manipulated variables MV and D is the number of disturbance variables DV). The control prediction process model 70 produces on an output line 72 a previously calculated prediction for each of the control and auxiliary variables CV and AV and a vector summer 74 subtracts these predicted Values for the current time from the actual measured values of the control and auxiliary variables CV and AV to produce an error or correction vector on an input 76.
The control prediction process model 70 then uses an N by M+D step response matrix (which may include a time response of each of the N process outputs to each of the M+D process inputs in response to a step change in the process inputs) to predict a future control parameter for each of the process outputs over the prediction horizon based on the disturbance and manipulated variables provided to other inputs of the control prediction process model 70. The control prediction process model 70, which is generally a linear process model having a transfer function that defines the response of each process output to each of the process inputs, also produces predicted steady state values of the process outputs at the end of the prediction horizon. The set of predicted values for each of the process outputs up to and including the prediction horizon is a prediction vector for that process output. However, because these prediction vectors are based on a linear or step response model, this prediction vector does not take into account or model any non-linear characteristics or relationships between particular process outputs and process inputs. In particular, the relationship between one of the process inputs and the process outputs may not be able to be accurately modeled using a linear or step response curve as is typically performed by the process model block 70. Thus, when one or more of the process outputs exhibits a non-linear relationship to one or more of the process inputs, the prediction vector developed by the process model block 70 is inaccurate due to these un-modeled non-linearities.
To compensate for such non-linearities, a correction or compensation unit 78 receives and uses the prediction signals 60 (which may each be a single steady state prediction value or a prediction vector having multiple prediction values at different times in the future) developed by the non-linear process model 42 and produces a compensated prediction vector that incorporates or accounts for the process non-linearities for each of the non-linear control and auxiliary variables (process outputs). It will be understood that the correction unit 78 may correct each control or auxiliary variable prediction vector (i.e., each process output prediction vector) separately using a different prediction signal 60 developed by the non-linear process model 42. Additionally, the correction unit 78 need only compensate the prediction vectors for process outputs that exhibit a non-linear relationship with one or more of the process inputs while leaving the prediction vectors as developed by the linear prediction process model 70 for each of the other linear process outputs unaltered. In this case, for example, the non-linear process model 42 may develop a separate or different prediction signal 60 for each of a sub-set of the control and auxiliary variables (process outputs used by MPC block 44 which exhibit a non-linear relationship with one or more of the process inputs (the manipulated or disturbance variables) and the correction unit 78 may compensate each of the prediction vectors for the process outputs within the subset separately while leaving the other process output prediction vectors as developed by the control prediction process model 70 unaltered.
The MPC block 44 also includes a control target block 80 that determines a control target vector for each of the N target control and auxiliary variables CVT and AVT (the target process outputs) provided thereto using a trajectory filter 82 previously established for the block 38. In particular, the trajectory filter 82 may provide a unit vector defining the manner in which control and auxiliary variables are to be driven to their target values over time. The control target block 80 uses this unit vector and the target variables CVT and AVT to produce a target vector for each of the control and auxiliary variables (process outputs) defining the changes in the target variables CVT and AVT over the time period defined by the prediction horizon time. A vector summer 84 then subtracts the compensated prediction vector for each of the control and auxiliary variables CV and AV (the process outputs) from the target vectors to define an error vector for each of the control and auxiliary variables CV and AV (the process outputs). The error vector for each of the control and auxiliary variables CV and AV is then provided to an MPC algorithm in a block 86 which operates in a standard manner to select the manipulated variable MV steps that minimize, for example, the least squared error, over the prediction horizon. Of course, the MPC algorithm block 86 may use an M by M process model or control matrix developed from relationships between the N control and auxiliary variables input to the MPC block 44 and the M manipulated variables output by the MPC block 44. These relationships are generally determined by inverting the control matrix used by the block 70, which inversion process is time consuming and difficult to regenerate during operation of the process 50.
Of course, the MPC algorithm block 86 may operate in any known manner to determine a set of manipulated variables to be delivered to the process 50 to drive the process outputs (the control and auxiliary variables CV and AV) to the desired set points CVT and AVT. While the control block 86 is illustrated as including a square control matrix developed from a square process model, this need not be the case and, in fact a non-square control matrix can be used if so desired, it being understood that such a non-square control matrix is generally harder to determine due to the complex mathematical procedures needed to develop such a non-square control matrix. However, the control matrix used by the block 86 is generally developed from a linear process model, i.e., the one used in the block 70 and, thus, does not produce the exact manipulated or control signals that will drive the process outputs of the process 50 to the desired set points in the fastest manner. However, because the error signals developed and provided to the MPC algorithm block 86 do account for the non-linear nature of the process outputs because of the correction unit 78, the MPC algorithm block 86 will drive the process outputs to the desired set points in a manner that is generally acceptable.
Furthermore, if desired, the MPC block 44 may include or be run with an optimizer which optimizes the operation of the MPC block 44. One such optimizer is described in detail in U.S. patent application Ser. No. 10/241,350, entitled “Integrated Model Predictive Control and Optimization within a Process Control System,” filed on Sep. 11, 2002, and published as Pub. No. US-2004-0049299-A1 on Mar. 11, 2004, which is assigned to the assignee hereof and the disclosure of which is hereby expressly incorporated by reference herein. Thus, it will be noted that the MPC block 44 may operate or function in any desired or known manner but incorporate the correction unit 78 to correct a predicted process output vector developed using a linear process model with a prediction signal developed by a non-linear process model or otherwise highly accurate process model, before such prediction vector is used to create one or more manipulated (control) signals to be used to control the process 50.
In particular, if the prediction signal 60 from the non-linear process model 42 is a prediction vector, the correction unit 78 may simply replace the prediction vector 96 developed by linear modeling with the prediction signal 60 so that the prediction vector 98 is essentially the prediction signal 60 developed by non-linear process model 42. Such a non-linear prediction vector may be produced with a dynamic neural network process model (i.e., a non-linear process model). However dynamic neural network modeling is difficult to develop and is not commonly applied. If, on the other hand, the prediction signal 60 from the non-linear process model 42 is a steady state signal indicating the predicted steady state of the process output CV at the control time horizon assuming the current inputs to the process 50 remain the same, then the non-linear process model prediction signal 60 may be used to adjust the prediction vector 96 to produce the prediction vector 98 using any desired technique. It should be noted that, if the non-linear process model 42 is a neural network process model, the steady state value of the process output CV can be produced or developed by freezing the current neural network model inputs and developing the neural network model output at the end of the process time to steady state (e.g., control time horizon).
One method of modifying the prediction vector 96 developed by the linear process model 70 using the prediction signal 60 developed by the non-linear process model 42 is to set the value of the prediction vector 96 at the end of the prediction horizon equal to the predicted steady state value 60 and then to proportionally modify each of the other values of the prediction vector 96 based on the time of each such point. If ΔSS is a correction at the end of prediction horizon at the scan k+1, then correction in any point i in the future can be expressed as:
s is the scan period;
i is the scan number in the future (i is equal to or between 0 and p−1); and
τ is the assumed process output time constant, which may be taken as the largest step response time constant for the corrected output. If desired, τ may be a fraction of the time to steady state, such the time to steady state divided by 3 or 4.
Based on this equation, Δ
nnk+1(p−1) is future (steady state) output value developed by non-linear process model at the scan k+1;
yk+1(p−1) is the predicted value of the process output p−1 scans in the future as developed by the linear process model at the scan k+1; and
The prediction vector
It will be understood from these equations that the prediction vector to be corrected for each of the process outputs includes a steady state value at a time horizon and that the correction modifies the prediction vector produced by the linear process model (such as in the MPC controller block) by adding to that prediction vector a difference between the predictions at the end of time horizon developed by the linear and the non-linear models multiplied by a first or higher order exponential function. In some embodiments, the first or higher order exponential function may use time constants equal to the process output time to steady state or, alternatively, may use time constants equal to some fraction of the process output time to steady state.
Alternatively, if process time constants are identified with a significant error, the prediction vector 96 can be corrected using process time to steady state
where TSS is time for the process steady state and 3≦α≦5
In general, higher order functions can be applied for developing a correction over the prediction horizon:
Δn k+1(i)=Δnn k+1(p−1)f(i,s,TSS)
where f (i,s,TSS) is a general exponential function higher than the first order that is used for prediction correction. As a general rule, it is advisable for the prediction correction to be the same order function as used for linear step responses for the considered output.
Each of these methods sets the intermediate prediction values (between the present time k and the time horizon k+p) in a way that preserves the dynamics of the response which, in turn, makes the modeling results compatible with Wiener and Hammerstein modeling. Of course, if desired, other manners of correcting the prediction vector developed by the linear process model 70 with the prediction signal 60 developed by the non-linear process model 42 can be used as well or instead of those described herein.
It will be understood that the method of correcting a model prediction vector developed using a linear process model with the output of a non-linear process model to compensate for non-linearities within a process as described herein is relatively easy and simple to implement during operation of the process without the need of a lot of computational processing to regenerate controller gains or non-linear model variables. It will also be understood that, because the MPC block 44 is still using a linear process model, the control outputs are not exactly configured to optimally drive the process 50 so that the process outputs reach the desired target values. However, because the process output prediction vectors are updated using the non-linear process model 42 after each scan, or periodically throughout operation of the process, i.e., after each n scans (where n can be any desired number), the MPC control matrix 86 will produce control signals that cause the process outputs to converge to the desired or target values relatively quickly despite the non-linearities within the process 50.
It will also be understood that, for the prediction signal 60 developed by the non-linear process model 42 to be applied correctly, this prediction signal should be developed for the same set of inputs used to develop the prediction vector at the output of the linear model block 70. Assuring that the non-linear process model 42 and the MPC block operation are synchronized is an important detail and care should be taken so that the future or predicted outputs of these blocks are not shifted one scan with respect to one another. Thus, it is preferred that the block execution order includes executing the non-linear process model 42 before executing the MPC block 44 so that the prediction signal 60 is available when the process model 70 for the MPC block 44 is executed.
If the non-linear process model 42 is a neural network process model, filtering for neural network future prediction can be applied in a similar way as for the prediction shift. In particular, the modeling linear equation could be modified in the following way:
y 0 =C
y 0 =C
A, B, C are process model matrices
0≦F≦1 is the filter for unmeasured disturbances correction and
0≦f≦1 is the filter for neural network correction.
Of course, modifications of the above equations can be applied as well.
As will be understood, the MPC block 44 uses each of the prediction signals 106 and 108 to separately correct or compensate different prediction vectors within the MPC block 44 in the manner generally described above with respect to the signal 60 in the embodiment of
Furthermore, it will be understood that the technique of using a prediction signal developed by a highly accurate or non-linear process model to correct or compensate a prediction vector produced by a linear process model to compensate for process non-linearities may be used in any other controller component besides a controller routine and an optimizer routine, as specifically described herein.
While the advanced control blocks 38, 100, 109 and 120 have been illustrated herein as a separate function block having an MPC block 44 or 128 and one or more separate process model blocks 42, 102, 104, 110 122, the advanced control block 38 maybe implemented simply by communicatively interconnecting an MPC block and one or more process model blocks, each of which could be separate function blocks. Alternatively, if desired, the process models 42, 102, 104, 110 and 122 described herein may be incorporated into or be a part of the MPC block 44 or 128. Furthermore, while the advanced control blocks or systems 38, 100, 109 and 120 are described herein as including an MPC block as a multiple-input/multiple-output control block, other multiple-input/multiple-output control blocks may be used instead, such as fuzzy logic control blocks, optimizer blocks or any other control routines or part of control routines that use a linear process model to develop control signals for a process.
As will be understood, the MPC or advanced control logic routines and methods described herein enable a user to create advanced control blocks such as MPC control blocks that can be used in processes having non-linear properties without having to integrate a non-linear process model within the standard linear MPC logic. Thus, this system enables known and relatively simple to implement non-linear process models, such as neural network process models, to be used to correct a linear MPC operation to provide better control in non-linear processes with relatively small effort, simple implementation within a process control routine, flexibility and minimal intrusion into the MPC control logic.
While the advanced control blocks have been described herein as being used in conjunction with Fieldbus and standard 4-20 ma devices, they can, of course, be implemented using any other process control communication protocol or programming environment and may be used with any other types of devices, function blocks or controllers. Moreover, it is noted that the use of the expression “function block” herein is not limited to what the Fieldbus protocol or the DeltaV controller protocol identifies as a function block but, instead, includes any other desired type of block, program, hardware, firmware, etc., associated with any type of control system and/or communication protocol that can be used to implement some process control function. Also, while function blocks typically take the form of objects within an object oriented programming environment, this need not be case.
Although the advanced control blocks and associated MPC blocks and non-linear process models described herein are preferably implemented in software, they may be implemented in hardware, firmware, etc., and may be executed by any other processor associated with a process control system. Thus, the routines 38, 42, 44, 100, 102, 104, 109, and 110 described herein may be implemented in a standard multi-purpose CPU or on specifically designed hardware or firmware such as, for example, ASICs, if so desired. When implemented in software, the software may be stored in any computer readable memory such as on a magnetic disk, a laser disk, an optical disk, or other storage medium, in a RAM or ROM of a computer or processor, etc. Likewise, this software may be delivered to a user or to a process control system via any known or desired delivery method including, for example, on a computer readable disk or other transportable computer storage mechanism or modulated over a communication channel such as a telephone line, the internet, etc. (which is viewed as being the same as or interchangeable with providing such software via a transportable storage medium). Additionally, different portions of this software may be separately implemented on different processors and/or within different devices, if so desired.
Thus, while the present invention has been described with reference to specific examples, which are intended to be illustrative only and not to be limiting of the invention, it will be apparent to those of ordinary skill in the art that changes, additions or deletions may be made to the disclosed embodiments without departing from the spirit and scope of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5353207 *||Jun 10, 1992||Oct 4, 1994||Pavilion Technologies, Inc.||Residual activation neural network|
|US5519605 *||Oct 24, 1994||May 21, 1996||Olin Corporation||Model predictive control apparatus and method|
|US5568378 *||Oct 24, 1994||Oct 22, 1996||Fisher-Rosemount Systems, Inc.||Variable horizon predictor for controlling dead time dominant processes, multivariable interactive processes, and processes with time variant dynamics|
|US5841652 *||Apr 3, 1996||Nov 24, 1998||Scap Europa, S.A.||Adaptive-predictive control and optimization system|
|US5933345||May 6, 1996||Aug 3, 1999||Pavilion Technologies, Inc.||Method and apparatus for dynamic and steady state modeling over a desired path between two end points|
|US6047221||Oct 3, 1997||Apr 4, 2000||Pavilion Technologies, Inc.||Method for steady-state identification based upon identified dynamics|
|US6278899||Oct 6, 1998||Aug 21, 2001||Pavilion Technologies, Inc.||Method for on-line optimization of a plant|
|US6381504 *||Dec 31, 1998||Apr 30, 2002||Pavilion Technologies, Inc.||Method for optimizing a plant with multiple inputs|
|US6445963||Oct 4, 1999||Sep 3, 2002||Fisher Rosemount Systems, Inc.||Integrated advanced control blocks in process control systems|
|US6487459||Feb 16, 1999||Nov 26, 2002||Pavilion Technologies, Inc.||Method and apparatus for modeling dynamic and steady-state processes for prediction, control and optimization|
|US6587744 *||Jun 20, 2000||Jul 1, 2003||Brooks Automation, Inc.||Run-to-run controller for use in microelectronic fabrication|
|US6757570 *||May 31, 2000||Jun 29, 2004||Guided Systems Technologies, Inc.||System and method for adaptive control of uncertain nonlinear processes|
|US6901300 *||Feb 7, 2002||May 31, 2005||Fisher-Rosemount Systems, Inc..||Adaptation of advanced process control blocks in response to variable process delay|
|US7039475 *||Dec 9, 2003||May 2, 2006||Pavilion Technologies, Inc.||System and method of adaptive control of processes with varying dynamics|
|US20040049299 *||Sep 11, 2002||Mar 11, 2004||Wilhelm Wojsznis||Integrated model predictive control and optimization within a process control system|
|US20040117040 *||Dec 9, 2003||Jun 17, 2004||Bijan Sayyarrodsari||System and method of adaptive control of processes with varying dynamics|
|US20050149209 *||Dec 2, 2004||Jul 7, 2005||Fisher-Rosemount Systems, Inc.||Adaptive multivariable process controller using model switching and attribute interpolation|
|US20070088448 *||Oct 19, 2005||Apr 19, 2007||Honeywell International Inc.||Predictive correlation model system|
|1||Aspentech, "Aspen IQ(TM)-Inferred Quality for Advanced Control and Monitoring," Product Brochure, date unknown.|
|2||Jan M. Maciejowski, "Model Predictive Control with Constraints," Pearson Education Limited, Appendix A, Appendix B, 2002.|
|3||Piche, Stephen, Pavilion Technologies, "Nonlinear Model Predictive Control Using Neural Networks," UCSB Control Consortium Meeting, 1999.|
|4||Qin, et al., "An Overview of Industrial Model Predictive Control Technology", pp. 232-256 (1996).|
|5||Qin, S.J. and Badgwell, T.A., "An Overview of Nonlinear MPC Applications," NMPC Workshop, Ascona, Switzerland, Jun. 3-6, 1998.|
|6||U.S. Appl. No. 10/241,350, entitled "Integrated Model Predictive Control and Optimization in a Process Control System", filed Sep. 11, 2002.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7856280||Aug 2, 2006||Dec 21, 2010||Emerson Process Management Power & Water Solutions, Inc.||Process control and optimization technique using immunological concepts|
|US7856281||Nov 7, 2008||Dec 21, 2010||Fisher-Rosemount Systems, Inc.||On-line adaptive model predictive control in a process control system|
|US7930045 *||Oct 7, 2008||Apr 19, 2011||Emerson Process Management Power & Water Solutions, Inc.||Two-stage model predictive control technique|
|US7949417||Sep 22, 2006||May 24, 2011||Exxonmobil Research And Engineering Company||Model predictive controller solution analysis process|
|US7987145||Jul 26, 2011||Honeywell Internationa||Target trajectory generator for predictive control of nonlinear systems using extended Kalman filter|
|US8032235 *||Oct 4, 2011||Rockwell Automation Technologies, Inc.||Model predictive control system and method for reduction of steady state error|
|US8260732||Nov 24, 2009||Sep 4, 2012||King Fahd University Of Petroleum And Minerals||Method for identifying Hammerstein models|
|US8346711||Nov 24, 2009||Jan 1, 2013||King Fahd University Of Petroleum And Minerals||Method for identifying multi-input multi-output Hammerstein models|
|US8346712||Jan 1, 2013||King Fahd University Of Petroleum And Minerals||Method for identifying hammerstein models|
|US8509924||Dec 10, 2010||Aug 13, 2013||Emerson Process Management Power & Water Solutions, Inc.||Process control and optimization technique using immunological concepts|
|US8620631||Apr 11, 2011||Dec 31, 2013||King Fahd University Of Petroleum And Minerals||Method of identifying Hammerstein models with known nonlinearity structures using particle swarm optimization|
|US8706267||Oct 28, 2008||Apr 22, 2014||Fisher-Rosemount Systems, Inc.||Process model identification in a process control system|
|US9163828||Oct 31, 2011||Oct 20, 2015||Emerson Process Management Power & Water Solutions, Inc.||Model-based load demand control|
|US9217565||Feb 7, 2011||Dec 22, 2015||Emerson Process Management Power & Water Solutions, Inc.||Dynamic matrix control of steam temperature with prevention of saturated steam entry into superheater|
|US9235657||Mar 13, 2013||Jan 12, 2016||Johnson Controls Technology Company||System identification and model development|
|US9335042||Aug 16, 2010||May 10, 2016||Emerson Process Management Power & Water Solutions, Inc.||Steam temperature control using dynamic matrix control|
|US9436179||Mar 13, 2013||Sep 6, 2016||Johnson Controls Technology Company||Systems and methods for energy cost optimization in a building system|
|US9447963||Feb 7, 2011||Sep 20, 2016||Emerson Process Management Power & Water Solutions, Inc.||Dynamic tuning of dynamic matrix control of steam temperature|
|US20070225835 *||Feb 13, 2007||Sep 27, 2007||Yucai Zhu||Computer method and apparatus for adaptive model predictive control|
|US20070245770 *||Apr 18, 2007||Oct 25, 2007||Saudi Arabian Oil Company||Optimization of a dual refrigeration system natural gas liquid plant via empirical experimental method|
|US20080125881 *||Aug 2, 2006||May 29, 2008||Emerson Process Management Power & Water Solutions, Inc.||Process control and optimization technique using immunological concepts|
|US20090005889 *||Jun 27, 2008||Jan 1, 2009||Rockwell Automation Technologies, Inc.||Model predictive control system and method for reduction of steady state error|
|US20090062932 *||Oct 28, 2008||Mar 5, 2009||Fisher-Rosemount Systems, Inc||Process Model Identification in a Process Control System|
|US20090143872 *||Nov 7, 2008||Jun 4, 2009||Fisher-Rosemount Systems, Inc.||On-Line Adaptive Model Predictive Control in a Process Control System|
|US20090240480 *||Mar 19, 2008||Sep 24, 2009||Honeywell International Inc.||Target trajectory generator for predictive control of nonlinear systems using extended kalman filter|
|US20100087933 *||Oct 7, 2008||Apr 8, 2010||Emerson Process Management Power & Water Solutions, Inc.||Two-stage model predictive control technique|
|US20110125684 *||Nov 24, 2009||May 26, 2011||Hussain Al-Duwaish||Method for identifying multi-input multi-output hammerstein models|
|US20110125685 *||Nov 24, 2009||May 26, 2011||Rizvi Syed Z||Method for identifying Hammerstein models|
|US20110144772 *||Dec 10, 2010||Jun 16, 2011||Emerson Process Management Power & Water Solutions, Inc.||Process control and optimization technique using immunological concepts|
|U.S. Classification||700/29, 700/31, 700/53, 700/44, 700/28|
|International Classification||G05B23/02, G05B13/02, G05B13/04, G06N3/02|
|Cooperative Classification||G05B13/048, G05B13/027|
|European Classification||G05B13/02C1, G05B13/04D|
|Aug 25, 2003||AS||Assignment|
Owner name: FISHER-ROSEMOUNT SYSTEMS, INC., A DELAWARE CORPORA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:WOJSZNIS, WILHELM K.;BLEVINS, TERRENCE L.;MEHTA, ASHISH;REEL/FRAME:014413/0484
Effective date: 20030604
|Jun 3, 2008||CC||Certificate of correction|
|Feb 22, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Mar 18, 2015||FPAY||Fee payment|
Year of fee payment: 8